6-open putback request: 6948909 and 6948803

[Joe Darcy]

Hello Max.

On 05/21/10 02:29 AM, Weijun Wang wrote:
> Hi Joe
>
> Here are the webrevs:
>
>    http://cr.openjdk.java.net/~weijun/6948803/6-open/webrev.00/
>    http://cr.openjdk.java.net/~weijun/6948909/6-open/webrev.00/
>
> They are almost identical to the fix for openjdk-7 except for one thing: In 6948803, several certificates are needed in the test. In openjdk-7, we've enhanced keytool to generate certificates, so the test there generates the certificates at runtime. In openjdk-6, we cannot do this. So I've copied certificate and keystores files into the test. The keystore is a binary file, which you might find not comfortable to include.
>   
For test/sun/security/validator/CertReplace.java, I recommend using a 
different kind of InputStream to pass the certificate rather than a 
FileInputStream.  For example, a ByteInputStream would allow a byte 
array to hold the (now binary) contents of the certificate file 
test/sun/security/validator/certreplace.jks.

For test/sun/security/tools/jarsigner/diffend.sh, I strongly encourage 
this to be rewritten as a Java program.

For both webrevs, please manually make sure the copyright dates are correct.

Depending on sequencing, this change may need Oracle rather than Sun 
copyrights as Kelly's big copyright swap is out for review.

> BTW, I don't remember I've ever pushed into the openjdk-6 repo, except for SSR ones. Does this mean I needn't do that and code changes into 6uXX workspaces will get folded into openjdk-6 periodically? Is my current work necessary?
>   

There is no automatic porting from the 6 update train (or JDK 7 for that 
matter) into OpenJDK 6.  The current work is necessary.

Thanks,

-Joe

> Thanks
> Max
>
>
> On May 19, 2010, at 10:13 AM, Joe Darcy wrote:
>
>   
>> Hi Max.
>>
>> Thanks for thinking of OpenJDK 6 as one release to port your fixes :-)
>>
>> Weijun Wang wrote:
>>     
>>> Hi Joe
>>>
>>> These two bugs are just approved for 6u21 b05:
>>>
>>>   6948803 (P2) CertPath validation regression caused by SHA1 replacement root and MD2 disable feature
>>>  
>>>       
>> This fix sounds okay, but please post the webrev for review.
>>
>>     
>>>   6948909 (P3) Jarsigner removes MANIFEST.MF info for badly packages jar's
>>>  
>>>       
>> Since OpenJDK 6 should have inherited the fix for 6543940 from JDK 7, this regression would appear to affect OpenJDK 6 too.
>>
>>     
>>> Both are regressions of a previous 6uXX release. I hereby request them to be included in openjdk-6.
>>>
>>> Thanks
>>> Max
>>>  
>>>       
>> Please post the webrevs for this fixes.
>>
>> As Andrew has already commented on to the list, the general policy (http://openjdk.java.net/projects/jdk6/) for OpenJDK 6 includes "As a general guideline, if a change is applicable to both JDK 7 and OpenJDK 6, the change should be made in JDK 7 no later than the change is made in OpenJDK 6" so I trust you will also be making arrangements to get these fixes into JDK 7.
>>
>> Thanks,
>>
>> -Joe
>>
>>     
>
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/jdk6-dev/attachments/20100521/15a686b7/attachment.html